Artificial intelligence-assisted automatic and index-based microbial single-cell sorting system for One-Cell-One-Tube

Automated identification of impact spatters and fly spots with a residual neural network

In criminal investigations, distinguishing between impact spatters and fly spots presents a challenge due to their morphological similarities. Traditional methods of bloodstain pattern analysis (BPA) rely significantly on the expertise of professional examiners, which can result in limitations including low identification efficiency, high misjudgment rates, and susceptibility to external disturbances. To enhance the accuracy and scientific rigor of identifying impact spatters and fly spots, this study employed artificial intelligence techniques in image recognition and transfer learning. Two types of bloodstains obtained from simulation experiments were utilized as datasets, and a pre-trained neural network, ResNet-18, was employed for feature extraction. The original fully connected layer was replaced, and a new fully connected layer with a dimensionality of 2 was introduced to fulfil the task requirements. The results demonstrate that the transfer learning network model, based on ResNet-18, achieved a maximum accuracy of 93 % in morphologically identifying impact spatters and fly spots. The objective is to assist crime scene investigators and BPA analysts to identify bloodstains at homicide scenes conveniently, rapidly and accurately, thereby furnishing scientific evidence for scene reconstruction and advancing BPA toward intelligent practices.

Artificial Intelligence-Assisted Automatic Raman-Activated Cell Sorting (AI-RACS) System for Mining Specific Functional Microorganisms in the Microbiome

The microbiome represents the natural presence of microorganisms, and exploring, understanding, and leveraging its functions will bring about significant breakthroughs in life sciences and applications. Raman-activated cell sorting (RACS) enables the correlation of phenotype and genotype at the single-cell level, offering a solution to the bottleneck in microbial community functional analysis caused by challenges in cultivating diverse microorganisms. However, current labor-intensive manual procedures fall short in catering to the demands of single-cell functional analysis in microbial communities. To address this issue, we developed an artificial intelligence-assisted Raman-activated cell sorting system (AI-RACS) that integrates precise single-cell positioning, automated data collection, optical tweezers capture, and single-cell printing to elevate microbial single-cell RACS from manual to automated, validating the efficacy of the system by isolating aluminum-tolerant microbes from acidic soil microbiota. Leveraging the AI-RACS framework, we sorted 13 strains from red soil samples under near-in situ conditions, with all demonstrating strong aluminum tolerance. AI-RACS efficiently segregates microbial cells from intricate environmental samples, investigating their functional attributes and presenting a novel tool for microbial research and applications.

Optical-based microbubble for on-demand droplet release from static droplet array (SDA) for dispensing one droplet into one tube

Static droplet array (SDA) is a pivotal tool for high-capacity screening assays, yet extraction and collection the target droplets that contain unique analytes or cells from the SDA remains one major technical bottleneck that limits its broader application. Here we present an optical-based on-demand droplet release (OODR) system by incorporating a 1064 nm laser-responsive indium tin oxide (ITO) layer into a chamber array-based droplet microfluidic chip. By focusing the 1064 nm laser onto the ITO layer, microbubbles can be created via local heating to selectively push-out the droplets from the chamber. Then the released droplet is readily exported in a one-droplet-one-tube (ODOT) manner by the inherent capillary force into pipette tip. Releasing of the droplets containing fluorescein sodium demonstrated ∼100% successful rate (9 out of 6400 droplets were successfully released) and low residual (only ∼5% of the droplet volume remains in the chamber). White or fluorescence image-based releasing of single-cell-droplets directly after cell loading or multi-cells-droplets derived from on-chip single-cell cultivation for both E. coli and yeast cells further demonstrated the wide applicability of OODR. The present system is user-friendly and has the potential to be applied in various high-throughput screening assays, including single molecule/cell analysis, drug screening, and phenotype-based cell sorting.

Single-cell rapid identification, in situ viability and vitality profiling, and genome-based source-tracking for probiotics products

Rapid expansion of the probiotics industry demands fast, sensitive, comprehensive, and low-cost strategies for quality assessment. Here, we introduce a culture-free, one-cell-resolution, phenome-genome-combined strategy called Single-Cell Identification, Viability and Vitality tests, and Source-tracking (SCIVVS). For each cell directly extracted from the product, the fingerprint region of D2O-probed single-cell Raman spectrum (SCRS) enables species-level identification with 93% accuracy, based on a reference SCRS database from 21 statutory probiotic species, whereas the C-D band accurately quantifies viability, metabolic vitality plus their intercellular heterogeneity. For source-tracking, single-cell Raman-activated Cell Sorting and Sequencing can proceed, producing indexed, precisely one-cell-based genome assemblies that can reach ~99.40% genome-wide coverage. Finally, we validated an integrated SCIVVS workflow with automated SCRS acquisition where the whole process except sequencing takes just 5 h. As it is >20-fold faster, >10-time cheaper, vitality-revealing, heterogeneity-resolving, and automation-prone, SCIVVS is a new technological and data framework for quality assessment of live-cell products.